Previously, I have written implementation codes of Gaussian Process regression in Python and Matlab. However, for better interaction and visualisation, I am currently working on implementation in Mathematica, but I am really a beginner in Mathematica.
Different from Python and Matlab, Mathematica is not using something like linspace in plotting functions. Instead, we enter the range to Plot such that it can generate continuous function itself. This makes me confused about how to generate sample functions given Gaussian Process prior. Let say my prior is distributed with zero mean and Squared Exponential kernel. I have
seKern[x1_,x2_,params_List]:=params[[1]]*Exp[-(Norm[x1-x2]^2/(2*params[[2]]))]
Knowing that $$\mathbf{f}\sim\mathcal{GP}(0,\,k(\mathbf{x},\,\mathbf{x}'))$$ I am trying to sample the functions from
RandomVariate[MultinormalDistribution[0, seKern[i, j, {1, 20}]], 1]
But, here, I don't know how to express this sample function in Plot such that I can generate functions as